Reviewed by George L. Huttar, SIL

This book aims ‘to provide a systematic analysis of contact-induced language change between two reasonably well-documented, genetically unrelated and typologically different language families north of the Amazon—North Arawak and Tucanoan’ (2). It succeeds well, thereby answering Aikhenvald’s call for in-depth studies as a basis for inductive generalizations that can be made to distinguish (the kinds of) interlanguage similarities ‘due to genetic inheritance and those due to borrowing’ (1).

Four situations in the Vaupés area of Colombia and Brazil are described, along with their effects on the languages involved, within a typology of language contact: (i) several East Tucanoan languages and N. Arawak Tariana (multilateral areal diffusion, no diglossia, no dominance), (ii) Central Tucanoan Retuarã and N. Arawak Yucuna (two languages, no diglossia, no dominance), (iii) E. Tucanoan Tucano and Tariana (two languages, no diglossia, dominance), and (iv) Portuguese and Tariana (two languages, diglossia, possible incipient dominance). Most of the book provides a detailed account of (iii), but the other situations are treated sufficiently to provide a valuable basis for comparison that will interest historical linguists, sociologists of language, and, to some extent, creolists.

After Ch. 1 (1–31) lays out the questions to be addressed and sketches the general language contact setting, Chs. 2–6 describe in detail the effects of Tucano on Tariana (with brief reference also to Retuarã and Yucuna) phonology (33–57), typological profile and pronominal systems (59–76), nominal categories (77–111), verbal categories (113–51), and syntax and (briefly) discourse (153–73). While the main impact is the pervasiveness of E. Tucanoan structures throughout Tariana (in contrast with genetically and geographically close Baniwa), the reader also gets a good introduction to Tucanoan and N. Arawak structures, particularly in morphosyntax.

Ch. 7 (175–86) describes the effects of Portuguese on Tariana, and Ch. 8 (187–211) examines code-switching and code-mixing of four sorts: Tariana with E. Tucanoan languages, with Baniwa, and with Portuguese; and the mixing of Tariana dialects. Even more than elsewhere in the book, the strong value prohibiting any dilution of one’s Tariana, and the ridicule meeting any violations, come through very clearly. But also clearly presented are situations where mixing is more allowable—such as the use of ‘a language distinct from that of the narrative’ (191) for the speech of nonhumans, and recourse to Portuguese to refer to ‘white man’s things’.

In Ch. 9 (213–21), A’s description of linguistic awareness among the Tariana lays a further foundation for the generalizations about language contact and change presented in subsequent chapters. Ch. 10 (223–41) summarizes the kinds of indirect (structural—by far the most prevalent) and direct (formal) diffusion in the Vaupés, while Ch. 11 (243–64) relates the changes in Tariana to the language’s obsolescence.

The concluding chapter (265–79) provides a comparison of the kinds of changes observed in the four sociolinguistically defined kinds of language contact examined, with generalizations and hypotheses that should serve well further exploration of what kinds of language features are most readily diffused in what kinds of contact situations.

Five appendices (281–325) on language classification, main features of languages of the area, and language proficiency data in one Tariana village, followed by a list of some 400 references and three indexes, conclude the book. There are also sixteen photographs and one map. Overall, style and organization are lucid, although typos, occasional misglossing, and other editorial infelicities are surprisingly many, but seldom a serious impediment to clarity.

Reviewed by Fredrik Heinat, Sweden

This book explores the lexicon-syntax interface. One of the author’s basic assumptions is that syntactic argument structure is projected as a function of individual lexical items. Martin Haiden applies theta theory (TT; Reinhart and Siloni 2005) to a range of constructions in German. Ch. 1, ‘Lexical semantics and cognition’, contains a general background to the problem of linking arguments to thematic roles. Following a brief description of TT, H tries to tie its features to more general concepts (e.g. mental activity [m] and physical activity [c]). These concepts are grounded in a commonsense theory of physics and a commonsense theory of mind mechanisms (37).

In Ch. 2, ‘German verbs’, H applies TT to German verbs. Investigating arguments in transitive, anti-causative, and reflexive constructions, H concludes that TT’s predictions are mostly accurate. However, he suggests three (rather than TT’s two) contexts that allow nonrealizations of thematic roles: (i) underspecified clusters can be assigned to external—but not internal—arguments, (ii) the presence of both [+c] and [-c] allows for nonrealization of the internal argument, and (iii) the nonrealization of a [+m] cluster is possible if arguments are [+m], [-c, -m], or [-m].

Ch. 3, ‘A bare phrase structure of argument expressions’, develops a theory of syntactic projection. H claims that thematic roles are provided by lexical items and then merged with syntactic heads of the category noun or verb (V) and that the roles are assigned to syntactic categories of Vs (196). This means that thematic roles are treated as independent syntactic objects. This analysis also incorporates morphophonological aspects of causativization and derivational morphology.

Ch. 4, ‘Applications and extensions’, adopts this analysis to for infinitival complements such as present participles, modal infinitival complements, and other participial constructions. H claims that it is the infinitival suffix that checks the features of V, and, under certain circumstances, it can also realize the internal theta-role. Another claim is that intransitive manner of motion verbs are lexically derived reflexives. The participial suffix plays the same role as the infinitival suffix: it checks the features of V and can sometimes realize V’s internal theta-role.

This well-structured and well-written book applies a theoretical framework (TT) to a wide range of different constructions in one particular language (German) in a thorough and systematic way. However, given the changes H had to implement to describe German, it would be interesting to see how TT—which seems to be based mostly on English data—accounts for less-closely related languages: one might suspect that things are not as neat as this theory implies.

References

Reinhart, Tanya, and Tal Siloni. 2005. The lexicon-syntax parameter: Reflexivization and other arity operations, Linguistic Inquiry 36.3.389–436.

Reviewed by Harald Hammarström, Chalmers University

The Tamang number over a million and are Nepal’s fifth largest language group. The dialect of Dhankute has considerably fewer speakers, around 10,000 to judge from census figures, and almost all are bilingual in Nepali. The Dhankute dialect has not been previously described, but there are dictionaries and grammar sketches of other Tamang varieties (by Westerners as well as natives). The present work appears to be based on interviews with native speakers, but the location is not specified.

Tamang belongs to the Tamang-Gurung-Thakali-Manang subgroup of Tibeto-Burman. The debates concerning the relationships of the various Tibeto-Burman subgroups are not entered into and, though a few passages from ethnographic works are cited, the contribution of this work is a synchronic sketch of the Dhankute variety.

The book sets off with a fairly extensive phonological description. There are ten vowels, eight diphthongs, and thirty-seven consonants reflecting the fact that all consonants (except h) have an aspirated counterpart. Stress is not phonemic and there is no phonemic tone. The absence of tones is quite interesting given that other varieties of Tamang have been described as having four tones (cf. Matisoff, J. A. Genetic versus contact relationship: Prosodic diffusibility in South-East Asian languages. Areal diffusion and genetic inheritance, ed. by A. Aikhenvald and R. M. W. Dixon, 291–327. Oxford University Press, 2001; Noonan, M. Chantyal. The Sino-Tibetan languages, ed. by G. Thurgood and R. J. LaPolla, 315–335. Routledge, 2003).

Following the phonology section, there is a lexicon section that contains notes on the lexical make-up and information on morphophonological processes. The next chapter, ‘Syntax’, presents syntax and morphology. Although the order of presentation is not always what one would expect, all major morphosynctactic aspects—word classes, phrases, clause combinations, and the tense/aspect system—are covered. On all of these matters Dhankute Tamang looks like a typical Tibeto-Burman language of this region. The demonstratives (and locative adverbs) encode not only distance but also three levels of physical location, that is, low ‘this-down-here’, level ‘this-at-the-same-level-as-us’, high ‘this-up-here’.

The grammar should be of interest to Himalayists and typologists alike. The author is committed to a modern functional description style with interlinear examples. There is a more than fair amount of spelling, typesetting, and reference errors but nothing serious.

Reviewed by Petra Gieselmann, University of Karlsruhe

Question answering (QA) is an upcoming research topic: the rise of the internet and the success of document retrieval has generated customer demand for good QA engines. Intelligent technologies are therefore used so that modern QA systems can combine complex natural language processing techniques, sophisticated linguistic representations, and advanced machine learning methods to find exact answers to a variety of natural language questions. The eighteen papers in this volume focus on present QA research and provide an extensive overview of this field. The volume is divided into six sections, which address different viewpoints and various approaches.

Part 1 introduces various QA approaches: the opening chapter by Dan Moldovan, Marius Paşca, and Mihai Surdeanu presents the Language Computer Corporation’s QA system, which was one of the most successful participants in the Text Retrieval Conference evaluation (TREC; an annual event sponsored by the United States National Institute of Standards and Technology). Relying on deep linguistic processing combined with logic theorem, the authors illustrate how multiple resources can be used to support a wide range of questions. In ‘A statistical approach for open-domain question answering’, Abraham Ittycheriah describes a completely different approach that features machine learning techniques such as maximum entropy modeling. Jose Vicedo and Antonio Ferrández present a detailed analysis of the problem of coreference resolution and its impact on QA.

Part 2 explores question processing. In ‘Questions and intentions’, Sanda Harabagiu explains the difficulties of question understanding, which includes the implications of user intentions. Using metonymies and the implications of user intentions to generate predictive and implied questions may reveal ways to enhance the efficiency, efficacy, and user satisfaction of QA results. Tomek Strzalkowski, Sharon Small, Hilda Hardy, Paul Kantor, Wu Min, Sean Ryan, Nobuyuki Shimizu, Liu Ting, Nina Wacholder, and Boris Yamron present their high-quality interactive QA, in which the user and the system first negotiate the scope and shape of the information and then cooperate in finding the answer within a large repository of unstructured data. Finally, Brigitte Grau, Olivier Ferret, Martine Hurault-Plantet, Christian Jacquemin, Laura Monceaux, Isabelle Robba, and Anne Vilnat examine linguistic variation within questions. Since paraphrases can take place either at the term or sentence level, an extensive use of natural language processing (NLP) components is essential.

Part 3 focuses on QA as a kind of information retrieval. The approach taken by Laszlo Grunfeld and Kui-Lam Kwok uses pattern matching as well as some information retrieval techniques, which work well for factoid questions, even without a deep linguistic analysis. However, Grunfeld and Kwok’s technique is inferior to sophisticated systems with knowledge inference and deep NLP. Failures are often due to the fact that answers are part of longer sentences and cannot be extracted correctly. Charles Clarke, Gordon Cormack, Thomas Lynam, and Egidio Terra present their passage selection approach to QA: the MultiText QA system first retrieves passages within a corpus that are related to the topic of the question and might, therefore, contain the answer. Then an answer is selected out of these passages by considering answer type and candidate redundancy. In ‘Query modulation for web-based QA’, Dragomir Radev, Hong Qi, Zhiping Zheng, Sasha Blair-Goldensohn, Zhu Zhang, Weiguo Fan, and John Prager evaluate QA using statistical models, which learn the best transformations to paraphrase a natural language question in preparation for a query to be sent to a search engine.

Part 4 deals with answer extraction. John Prager, Jennifer Chu-Carroll, Eric Brown, and Krzysztof Czuba present their approach to QA through predictive annotation: they index a database with extended named entities that might become answers in factoid questions. The next chapter by Rohini Srihari, Wei Li, and Xiaoge Li describes how information extraction can be used to support QA. Using multiple levels of information extraction, they generate a precise answer to a factoid question. Finally, Abdessamad Echihabi, Ulf Hermjakob, Eduard Hovy, Daniel Marcu, Eric Melz, and Deepak Ravichandran address the issue of how to select an answer string. Evaluating knowledge-based, pattern-based, and statistics-based answer selection, they claim to improve the individual answer selection modules and combine their outputs in a maximum entropy model.

Part 5 explains methods and techniques employed to evaluate the performance of QA systems. Ellen Voorhees looks at the evaluation methods used within the QA track of TREC, the first large-scale evaluation of open-domain QA systems. William Hersh describes two evaluation initiatives: the interactive QA track on TREC and user studies in the medical domain. He argues that studies with users are essential for understanding the efficacy of QA systems. The last chapter of this section by William Ogden, James McDonald, Philip Bernick, and Roger Chadwick explores habitability in QA systems.

Finally, Part 6 deals with perspectives on QA. Steven Maiorano discusses QA as a technology for intelligence analysis. He believes that QA can help various analysts to get answers for specific questions in a dialog-based system. The next article by Ellen Riloff, Gideon Mann, and William Phillips explains the need for a large collection of question answer pairs to train statistical models for QA. Therefore, they developed a method to automatically generate question answer pairs from normal text. Finally Mark Maybury deals with new directions in QA. He discusses the need for answering broader and deeper questions, extracting answers from varied sources, and supporting users with different backgrounds.

This volume provides a good overview of current research in QA. It will be interesting for students just beginning to look at QA as well as for researchers wanting to gain insights into interactive QA.

Reviewed by Harald Hammarström, Chalmers University

The present volume is a collection of sixteen papers mainly stemming from a conference in 1989. Its appearance now, however, is very welcome given the renewed controversy over the applicability of the comparative method and the genetic relations of Australia’s languages. The volume concerns only non-Pama-Nyungan—the continent’s least known and linguistically most diverse set of languages—but not every non-Pama-Nyungan group is given detailed treatment.

The introductory chapter by the editor covers history, summarizes the contributions in the present volume, enumerates non-Pama-Nyungan groups not featured in the volume, and hints at the future, namely, the possibility of proving the bulk of non-Pama-Nyungan as a genetic unit based on pronouns resemblances. Subsequent chapters are family or subgroup surveys of varying depth. For example, diachronic typology of head-marking in Jarragan is discussed by McConvell; Southern Daly is demonstrated through articles by I. Green and Reid; Eastern Daly is demonstrated by Harvey (verb systems and object enclitics); and Garrwa and Wanyi are separate but genetically related languages as shown by Breen (with an update by Belfrage).

Nyulnyulan is demonstrated by Stokes and McGregor using the comparative method and lexicostatistics, though here one is left with some questions since the authors choose to call it a family-like unit (the meaning of this vs. ’family’ or ’genetic unit’ is never explained) despite their clear application of the comparative method. Also, illogically they argue (60) that lexicostatistic evidence is admissible if and only if it confirms the conclusions from the application of the comparative method (and if so, the conclusions reached from the comparative method are said to be strengthened!).

In my opinion, the biggest contributions are the demonstration of an extended Gunwinyguan family (also with an evidence-backed subgrouping) through a first proto-phonology by Harvey, and evidence from verb morphology by Harvey, Alpher, Evans, Merlan, and R. Green. Whereas Alpher, Evans, and Harvey want to include Mangarrayi on the counts of verbal morphology, Merlan convincingly argues that evidence from nominal prefixation gives precedence to an inclusion in Jeff Heath’s Maran (aka Marra-Alawa) family.

Finally, there is a chapter by Harvey on the tentative reconstruction of pronominals for a hypothetical proto non-Pama-Nyungan language spanning most but not all non-Pama-Nyungan languages.

All articles show a very high level of awareness of the problem of distinguishing diffusion vs. inheritance in historical linguistics. They also cite unpublished descriptive materials—not fieldnotes, AIATSIS documents, or soon-to-be-published drafts, but extensive typed-up grammar sketches and dictionaries that have apparently been circulated for ages among the initiated. It would be more beneficial if they were somehow made available (in whatever form). Sadly, this situation is seen in linguistic typology all too often. Indeed, if something is worth citing, it is foul play to fail to allow the whole scientific community to learn from it.

With its wealth of detailed, otherwise hard-to-find data this book is a must for researchers in Australian linguistics. It is also very valuable for those interested in historical/areal typology since most of the book discusses grammatical rather than lexical forms. World linguists will appreciate the hard evidence for genetic groupings of non-Pama-Nyungan languages, though it is sometimes a tough read for the nonexpert. The articles freely mention well-established families in the same manner as premature groupings (proto-Arnhem Land, proto-non-Pama-Nyungan, proto-Australian, family-level isolate (a lexicostatistics term), Gunwinyguan with/without Anindilyakwa, etc.) so one has to keep track of the evidence oneself.

Reviewed by Thomas R. Wier, University of Chicago

When pressed, most linguists will agree that the traditional arguments about grammatical categories have a long and confused history that explains lexical idiosyncrasies as exceptions to generalizations that do not take part in productive morphosyntactic processes. Etsuyo Yuasa’s book on modular mismatches in natural language is in part an answer to this problem. By adopting a multimodular approach along the lines of autolexical syntax (Sadock 1991) or that of Culicover and Jackendoff (2005), such cases of categorial mismatch are seen as the lifeblood of modular autonomy, as they constitute the actual direct evidence that distinguishes one grammatical module from another. Incorporating constructional approaches and multimodularism, Y’s work represents a unique and important contribution to the effect of lexical variation on syntactic and semantic typology.

In Chs. 1 and 2, Y introduces the basic notion of grammatical modularity and discusses several instances of how the mismatch between the modules illustrates modular autonomy. Furthermore, Y discusses how multimodular approaches formally distinguish prototypical and nonprototypical syntactic and semantic constructions. Following this theoretical foundation, Y presents a series of case-studies drawing primarily from English and Japanese data.

Ch. 3 examines English pseudo-coordinate conditionals and Japanese te-coordination. These two constructions are shown to be mirror-images of each other: English pseudo-coordinate conditionals are syntactically symmetrical and semantically asymmetrical, while te-coordination has some syntactic properties typical of Japanese subordinate clauses but semantically behaves like coordinated clauses. Y finds this unsurprising because it is a direct consequence of rigorously autonomous modules engendered by her theoretical framework.

Ch. 4 focuses on nonrestrictive relative clauses and demonstrates how nonrestrictive relative clauses differ from restrictive relative clauses in terms of scope relations, tense, adverbs, modal verbs, VP-deletion, and antecedent-contained deletion. In all respects, nonrestrictive relative clauses are more similar to independent clauses than to subordinate clauses. After discussing the shortcomings of previous analyses, Y provides her own account based on modular autonomy: that the two types of relative clauses, although syntactically similar, differ in their semantic properties because they are each generated independently, in a separate module, by separate generative rules.

In Ch. 5, Y looks at performative adverbial clauses, yet another kind of clausal mismatch. Performative adverbial clauses (He’s telling a lie, because he’s sweating) behave like independent clauses, not like subordinate clauses in terms of scope, tense, adverbial and modal modification, and VP-deletion. Once again, the explanation is modular autonomy: performative adverbial clauses are subordinate only in syntax but behave like fully independent clauses in semantics and pragmatics.

Ch. 6 surveys categorial mismatch. Traditional Japanese grammar has long identified the unusual properties of certain subordinating conjunctions that behave in some respects like complementizers, in other respects like regular nouns. For Y, this is mysterious only if one assumes rigorous isomorphism between modules. Because multimodular analyses do not make this assumption, they allow for precisely such mixed categories.

Y’s primary interest is the larger philosophical question for linguistics: what is the value of apparently exceptional data, and what should the response to these exceptional data be? One mode of thought moves all such data to the theoretical periphery—often of a poorly-defined status—much as the orbit of Mercury was inexplicable in physics until relativity came along. Another mode of thought takes exceptions at face value, accepting the need to abandon previous analyses in favor of something new. Y’s work makes a strong case that the latter attitude is appropriate for these English and Japanese data, thus making an important contribution to the scientific methodology of linguistics.

References

Culicover, Peter W. and Ray Jackendoff. 2005. Simpler syntax. Oxford: Oxford University Press.

Sadock, Jerrold M. 1991. Autolexical syntax: A theory of parallel grammatical representations. Chicago: University of Chicago Press.

Reviewed by Dinha T. Gorgis, Jadara University for Graduate Studies

Hall’s book includes four parts with a preface and a glossary. For H, ‘language isn’t just a way of transmitting meaning’ (xi), but also a way of transmitting ‘power’ (xii) through which we can see ourselves and understand the other. Although the book is about linguistics, yet it is not a conventional textbook; it ‘embraces and integrates the social and psychological aspects of language’ (xiii).

Part 1, ‘Magic’, has two chapters. Ch. 1 (3–27) is an invitation to finding out ways to break the language spell with which we live, linguists included. Thought is encoded in language, which is expressed through speech (or text), as its ‘external manifestations’ (6). We are put under the spell probably because language is invisible to us. And this ‘led our preliterate ancestors to associate it with magic’ (7). We all know that language is there, but we are unable to ‘penetrate its mystery’ (10). The paradox then is one of invisibility vs. presence. And this is what H calls the language spell. Three mental domains are assumed to encode the spell: linguistic knowledge (‘our language software’), linguistic awareness, and linguistic belief, the first being ‘most fundamental’ (14).

Ch. 2 (28–46) introduces the field of linguistics that H criticizes as not having been able to make the inner workings of language visible enough to break the spell. He therefore appeals to language professionals to abandon their ‘ivory tower’ (28) and invites them to do down-to-earth linguistics. H attempts to break the spell by demonstrating how linguistic knowledge operates (Fig. 1, p. 36), and how language use works (Fig. 2, p. 44).

The three chapters in Part 2 handle just ‘words’. In Ch. 3 (49–84), one finds the amazing world of words, names, meanings, and thought, and their associated problems. Fig. 1 (67) nicely depicts how our word information flows between the outside world and the human mind. Only then comes the question of what a word is. Having been aided by ample evidence from reported experiments, a word for H ought not to be envisaged as a bi-dimensional thing like a coin ‘but rather a “triad” of three things at least’ (79; see Fig. 4, p. 80).

Ch. 4 (85–109) continues ‘to challenge the public Spell-bound view of language’ (85). It seeks to find out where words come from, starting with language acquisition, going beyond the conventional dictionary, delving into the mental lexicon, and relating internalized language (I-language) to externalized language (E-language). Ch. 5 (110–29) deals with word forms and how words come into existence. H tries to draw distinctions between speech and spelling, between phonemes and graphemes, and, most importantly, between ‘phonetic procedural knowledge’ and ‘declarative knowledge of phonology’. He regrets that linguists are IPA spell-bound. Several factors seem to conspire in the creation of new words, such as crosslinguistic influence, code-switching, and borrowing ‘via contact’ (127).

Part 3, ‘Grammar’, includes three chapters. Ch. 6 (133–53) handles ‘morphology’. Combinatoriality within the word and between words is approached ‘from the perspective of I-linguistic declarative knowledge … and how such knowledge changes through time’ (133). Ch. 7 (154–86), ‘Syntax’, shows us that a sentence exhibits a hierarchical structure, just like a syllabic structure, and that when syntax is put into action, it interacts with pragmatics. Demonstrably, Fig. 15 (183) illustrates a small part of an English syntactic component, lexical component, and conceptual structure, which is assumed to reside in all English minds.

In Ch. 8, H moves from DNA to discourse community (187–208). The innatist view of language is challenged as being insufficient, though plausible. To go beyond syntactic rules, trees, and semantics, pragmatic knowledge and communicative competence must go together with our linguistic competence. It is this interface that can ‘resolve more of the reality behind the Language Spell’ (208).

Part 4, ‘Babel’, concludes the book with three more chapters. Ch. 9 (211–37) explores interlinguistic diversity. Although we share one human language emerging from our DNA, our E-languages are different. H notes that ‘one nation, one language’ is yet a second spell that we equally need to combat as long as we believe that ‘languages change rapidly and intermix freely, leading to the multilingual world’ (236) in which we live.

Ch. 10 (238–69) is a move from a macro to a micro view of surface diversity. Thus, H has extended Dell Hymes’s acronym SPEAKING, which is ‘essentially a product of external sociocultural forces, rather than internal linguistic ones’ (268). Ch. 11, ‘The spell unbroken’ (270–99), addresses the question of linguistics in action. The linguistic knowledge, awareness, and belief are discussed once more and related to some of the ways linguistics and allied fields can contribute to the breaking of the spell. ‘And yet the Spell will remain completely unbroken for most until linguists are heard and read more widely’ (297).

H’s book is unique in presentation, coverage and, above all, style; it is breathtaking. He attempts to convince us through an intellectual and uninterrupted journey that we are able and/or should be enabled to break the language spell. Except for Fig. 3 on p. 158, which unfortunately is missing, the whole book is error-free and elegantly presented. And passing such a judgment may be infected by the spell!

Reviewed by Karen Steffen Chung, National Taiwan University

In this dissertation-based study, Benczes asks and attempts to answer questions like how we can, in the right context, understand that land fishing means ‘metal detecting’, and has nothing to do with what we normally understand as fishing. The answer, B claims, lies in our extensive use of metaphor and metonymy when we create new noun compounds. This also explains why such compounds have often been considered exceptional or idiosyncratic in many previous analyses. B’s main thesis is that compounds like these are not exceptional but are in fact as analyzable as other compound types. Their main distinguishing feature is not a lack of transparency, but rather creativity, especially when compared to straightforward endocentric compounds such as apple tree. According to B, ‘creativity’ involves the use of novel figures of speech, particularly, in the case of noun-noun compounds, metaphor and metonymy, in contrast to the generativist definition of ‘creativity’ as a simple stringing together of the components of a sentence (7).

B compiled her database of seventy-eight novel noun-noun compounds from works such as the Longman Dictionary of Contemporary English and the online dictionary of neologisms, Word Spy. B says that speakers do not coin new noun-noun compounds to enlarge the existing vocabulary of their language, but to fill a specific communication need in an effective and economical way (30). Through this work the reader will experience the sheer joy and fun people have with their language when they coin a new compound. A representative example is bar-code hairstyle (110), which came into English from the Japanese; it refers to a comb-over on a balding man.

B surveys the views of a number of different writers on exactly what is meant by ‘exocentric’ compounds. Leonard Bloomfield includes metonymic (e.g. phone neck ‘a sore neck resulting from too much time on the phone’) and combined metaphorical-metonymic ones (e.g. bell-bottoms), but not purely metaphorical ones like beanpole. Because B includes purely metaphorical compounds, she rejects the term ‘exocentric’ altogether (183), feeling it comprises too inclusive a continuum of noun-noun compound types, and thus becomes difficult to define precisely. In my view, however, exocentric compounds can be unambiguously defined morphosyntactically: they are noun compounds in which the nominal head does not appear, for example, pickpocket ‘one who picks pockets’, and a carrot top ‘one who has red hair’. Beanpole, by contrast, identifies a person directly with a beanpole, and it does have a head, namely pole, although the compound as a whole uses a symbol to stand in for the referent, in this case, ‘a very thin person’.

There are a few dozen minor errata in the text, but otherwise the English is clear and flows well. B’s study is valuable both for its fresh approach and findings, and also for the contribution it makes to stimulating further research interest in morphology and compounding.

Reviewed by Colette M. van Kerckvoorde, Bard College at Simon’s Rock

The way in which we use language on a daily basis is not our own choice. Rather, our linguistic behavior is influenced and heavily determined by a small number of powerful members in our society who regulate and manipulate language to create a fertile ground for their ideologies to thrive. When people are told how to speak, basic democratic rights are violated. Sometimes, language policies are stated explicitly in official documents, but more frequently, policies can only be derived implicitly from a close study of de facto practices. In this volume, Elana Shohamy argues that language policies should be examined from a broad perspective that includes not only formal and declared policies but also mechanisms found in most societies to create desired language practices.

The book is divided into three parts. In Part 1, S indicates that language is free, dynamic, creative, personal, and without defined boundaries. This view runs counter to the widespread myth that language is stagnated, rule-bound, closed, and finite. This myth, S states, was fostered by the formation of the nation-states in Europe as well as the emergence of the scientific study of language. As the nation-states emerged, they began to define which language their citizens should use and thus made language into a symbol of loyalty and patriotism (i.e. status policy). At the same time, some linguists became interested in establishing the correct forms of the language (i.e. corpus policy).

Part 2 focuses on the various mechanisms that are used to encourage desired language practices in society. Four categories are distinguished: (i) official laws, rules, and regulations; (ii) language education, including second and foreign language education; (iii) language tests, especially those that determine entrance to educational institutions; and (iv) the linguistic landscape, such as street signs and advertising billboards. S demonstrates that most of these mechanisms should be considered policy devices because they create de facto policies. S frequently uses Israel to illustrate these different mechanisms, although occasionally she also draws on examples from the United States.

S demonstrates that the encouragement and imposition of certain behaviors has consequences. In Part 3, she summarizes the consequences of overt and covert language policies and proposes strategies to respond to violations of basic democratic rights. Such strategies emphasize the need for a democracy of inclusion: a view of nations based on ethnic unity is outdated. S further argues that people need to become aware of the hidden agendas behind certain language policies; she advocates that commonly accepted myths about language be debunked; and finally, she calls for language activism to change the current situation prevalent in many nation-states.

S raises interesting questions with regard to language policies and convincingly argues that many mechanisms violate basic human rights. Although her call for activism is appropriate, it remains vague. She neither provides any concrete examples of such activism, nor does she address how disadvantaged groups can find support for their fight.

This book will appeal to students from a variety of disciplines such as education, sociology, political science, philosophy, and communication studies. Linguists, too, will benefit from this book, since they especially ‘analyze and define the world in linguistic terms, often not realizing that languages represent only one aspect of identity’ (143).

Reviewed by Andreea S. Calude, The University of Auckland

There are few languages that encompass the history, diversity, and political load that can be observed with the Romani language. Spoken by the Roma, the Sinti, the Kale, and others most commonly placed under the label term of ‘gypsy’, Romani developed since the sixteenth century as a language of nomadic people, constantly on the move and often highly isolated from each other. From a linguistic standpoint, this has led to a rich source of interesting avenues for research, such as dialectal issues (relating to how the various dialects of the language evolved), language-contact problems (examining how the various dominant languages spoken in the countries that were settled by Romani speakers have influenced the Romani language), historical development, sociolinguistic concerns, and other general theoretical linguistics issues.

This volume—the proceedings to the 6th International Conference on Romani Linguistics held in Graz in 2002 (following previous conferences in Hamburg 1993, Amsterdam 1994, and Manchester 1998)—is divided into five parts: ‘Romani dialectology’ (Part 1), ‘Descriptive studies on individual Romani dialects’ (Part 2), ‘Language change with and without contact’ (Part 3), ‘Computational linguistics’ (Part 4), and ‘Sociolinguistics’ (Part 5). The computational linguistics section (though only containing one paper) is indicative of the innovative directions undertaken by Romani linguistics researchers, in using corpora and other such electronic databases.

Part 1 includes a discussion by Yaron Matras revisiting general issues of dialectology in Romani linguistics (7–22) and a study by Birgit Igla investigating the Sinti-Manuš dialect group (23–47).

There are four papers in Part 2. Lev Cherenkov gives a brief examination of the Russian variety of Plaščuno (63–47); Irene Sechidou analyzes the Greek Romani variety of Ajios Athanasios (48–59); Ignasi-Xavier Adiego looks at present day Caló (60–78); and Zoran Lapov discusses the different Croatian Roma dialects (79–89).

Part 3 includes three papers on verbal change phenomena: Desislava Draganova on the borrowing of Turkish verb forms into Bulgarian Romani (90–98), Barbara Schrammel on borrowing and calquing of German verbal affixes in Austrian Romani varieties (99–113), and Helena Pirttisaari on borrowed and inherited past participle morphs in Finnish Romani (114–27). The remaining two papers by Norbert Boretzky (128–43) and Gitte G. Simonsen (144–49) consider sounds change and semantic change, respectively.

Part 4 on computational linguistics contains one paper, by Kimmo Granqvist, who uses data from the ROMLEX project (documenting the Romani lexicon from as many different varieties and dialects as possible) to investigate the implementation of a two-level morphology processor for Finnish Romani (150–62).

The final part has three papers on different aspects of Romani sociolinguistics. Victor Friedman examines a trilingual Macedonian newspaper (163–73); Jelena Petrović and Lada Stefanović describe the situation of Roma refugees in Kosovo (174–81); and Elena Marushiakova and Vesselin Popov focus on the Gypsy nomadic groups of Bulgaria (182–87).

The volume constitutes a comprehensive and varied compendium of research conducted on Romani linguistics and is of interest to anyone working not only on Romani linguistics, but also on sociolinguistic and dialectal variation, language contact, as well as theoretical linguistics.